<\/p>\n
\u0130statistiklerderegresyon modellerini<\/a> iki nedenden dolay\u0131 kullan\u0131r\u0131z:<\/span><\/p>\n (1)<\/strong> Bir veya daha fazla a\u00e7\u0131klay\u0131c\u0131 de\u011fi\u015fken<\/a> ile bir yan\u0131t de\u011fi\u015fkeni<\/a> aras\u0131ndaki ili\u015fkiyi a\u00e7\u0131klay\u0131n<\/em> .<\/span><\/p>\n (2)<\/strong> Bir veya daha fazla a\u00e7\u0131klay\u0131c\u0131 de\u011fi\u015fkenin de\u011ferlerine dayanarak bir yan\u0131t de\u011fi\u015fkeninin de\u011ferlerini tahmin edin<\/em> .<\/span><\/p>\n Amac\u0131m\u0131z (2)<\/strong> bir yan\u0131t de\u011fi\u015fkeninin de\u011ferlerini tahmin etmek<\/em> oldu\u011funda, bunu yapmak i\u00e7in m\u00fcmk\u00fcn olan en iyi regresyon modelini kulland\u0131\u011f\u0131m\u0131zdan emin olmak isteriz.<\/span><\/p>\n Yeni veriler \u00fczerinde en iyi tahminleri yapacak regresyon modelini bulmak i\u00e7in kullanabilece\u011fimiz bir \u00f6l\u00e7\u00fcm, ” tahmin edilen<\/strong> art\u0131k<\/strong> kareler<\/strong> toplam\u0131<\/strong> ” anlam\u0131na gelen PRESS istatisti\u011fidir<\/strong> .<\/span><\/p>\n A\u015fa\u011f\u0131daki \u015fekilde hesaplan\u0131r:<\/span><\/p>\n BASIN = \u03a3(e i<\/sub> \/ (1-h ii<\/sub> )) 2<\/sup><\/strong><\/span><\/p>\n Alt\u0131n:<\/span><\/p>\n \u00c7oklu regresyon modelleri g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, en d\u00fc\u015f\u00fck PRESS’e sahip olan, yeni bir veri seti \u00fczerinde en iyi performans\u0131 g\u00f6sterecek olan olarak se\u00e7ilmelidir.<\/span><\/p>\n A\u015fa\u011f\u0131daki \u00f6rnek, R’deki \u00fc\u00e7 farkl\u0131 do\u011frusal regresyon modeli i\u00e7in PRESS istatisti\u011finin nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n \u00dc\u00e7 a\u00e7\u0131klay\u0131c\u0131 de\u011fi\u015fken ( x1<\/sub> , x2<\/sub> ve x3<\/sub> ) ve bir yan\u0131t de\u011fi\u015fkeni y’den olu\u015fan bir veri setimiz oldu\u011funu varsayal\u0131m:<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, lm()<\/strong> i\u015flevini kullanarak \u00fc\u00e7 farkl\u0131 regresyon modelinin bu veri k\u00fcmesine nas\u0131l s\u0131\u011fd\u0131r\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, her model i\u00e7in PRESS istatisti\u011finin nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n PRESS istatisti\u011fi en d\u00fc\u015f\u00fck olan modelin ise 519.6435<\/strong> PRESS istatisti\u011fiyle Model 2 oldu\u011fu ortaya \u00e7\u0131kt\u0131. Bu nedenle, yeni bir veri seti \u00fczerinde tahmin yapmak i\u00e7in en uygun modeli bu modeli se\u00e7ece\u011fiz.<\/span><\/p>\n Basit Do\u011frusal Regresyona Giri\u015f<\/a> \u0130statistiklerderegresyon modellerini iki nedenden dolay\u0131 kullan\u0131r\u0131z: (1) Bir veya daha fazla a\u00e7\u0131klay\u0131c\u0131 de\u011fi\u015fken ile bir yan\u0131t de\u011fi\u015fkeni aras\u0131ndaki ili\u015fkiyi a\u00e7\u0131klay\u0131n . (2) Bir veya daha fazla a\u00e7\u0131klay\u0131c\u0131 de\u011fi\u015fkenin de\u011ferlerine dayanarak bir yan\u0131t de\u011fi\u015fkeninin de\u011ferlerini tahmin edin . Amac\u0131m\u0131z (2) bir yan\u0131t de\u011fi\u015fkeninin de\u011ferlerini tahmin etmek oldu\u011funda, bunu yapmak i\u00e7in m\u00fcmk\u00fcn olan en iyi regresyon modelini […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[],"class_list":["post-1037","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\u00d6rnek: PRESS istatisti\u011finin hesaplanmas\u0131<\/strong><\/span><\/h3>\n
data <- data.frame<\/span> (x1 = c(2, 3, 3, 4, 4, 6, 8, 9, 9, 9),\n x2 = c(2, 2, 3, 3, 2, 3, 5, 6, 6, 7),\n x3 = c(12, 14, 14, 13, 8, 8, 9, 14, 11, 7),\n y = c(23, 24, 15, 9, 14, 17, 22, 26, 34, 35))\n<\/strong><\/pre>\n model1 <- lm(y~x1, data=data)\n\nmodel2 <- lm(y~x1+x2, data=data)\n\nmodel3 <- lm(y~x2+x3, data=data)\n<\/strong><\/pre>\n
#create custom function to calculate the PRESS statistic<\/span>\nPRESS <- function<\/span> (model) {\n i <- residuals<\/span> (model)\/(1 - lm. influence<\/span> (model)$ hat<\/span> )\n sum(i^2)\n}\n\n#calculate PRESS for model 1<\/span>\nPRESS(model1)\n\n[1] 590.2197\n\n#calculate PRESS for model 2<\/span>\nPRESS(model2)\n\n[1] 519.6435\n\n#calculate PRESS for model 3<\/span>\nPRESS(model3)\n\n[1] 537.7503\n<\/strong><\/pre>\n Ek kaynaklar<\/strong><\/span><\/h3>\n
Cimri bir model nedir?<\/a>
\u0130yi bir R-kare de\u011feri nedir?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"